Claude ID Verification, GLM 5.2 Self-Host, Codex UI Flip — Jun 21

435 messages · 62 active members

435
messages
62
active members
@anonymous, @Coybh, @jcartu
top contributors

Overview

Today's chat was dominated by Anthropic's announcement that Claude will require ID verification, sparking concerns about account farming, foreign access, and a domino effect across other model providers. anonymous surfaced the support article and @jcartu suggested the move stems from government pressure after Fable's capabilities 'scared the government,' predicting other labs will follow as they ship more powerful models. On infrastructure, @jcartu pushed hard on self-hosted inference, sharing a GLM-5.2-504B FullKD release and arguing 4-8 RTX 6000 Pro GPUs can replace Opus-tier coding for the cost of electricity (card prices already up from 750k to 850k rub in 3 months). Meanwhile a notable model preference flip emerged: several builders now prefer Codex/GPT-5.5 for UI and copywriting, reserving Claude Opus for strategy and creative writing. @tounano's standout workflow — generate UI with GPT-Image-2, then feed screenshots to Codex for ~90% fidelity code — is faster than prompting code directly and parallelizable across chat windows. Side threads covered agent orchestration stacks (Telegram ops agent + GitHub Issues/Jira + custom control center), Google Ads creative volume strategies (20-30 presell pages/month, PMax-to-DG promotion), Higgsfield's misleading 'Unlimited' marketing, deepfake celeb VSL ads inflating native CTRs, and AI trading harnesses where the eval harness matters more than the prompt.

Topics

anonymous surfaced Anthropic's new identity verification policy, prompting debate about account farming workarounds and worries it won'anonymous be limited to US nationals. @jcartu and anonymous speculated the move is downstream of government pressure after Fable's capabilities scared regulators, predicting a domino effect across other providers.

@jcartu shared a HuggingFace release of GLM-5.2-504B-FullKD and made a case for buying 4-8 RTX 6000 Pros to self-host: ~0 KLD loss, 8 concurrent sessions at 60-80tps, near Opus 4.8 coding quality. Card prices jumped 750k→850k rub in 3 months. Tradeoffs: NVFP4 quant degradation, 120k ctx at 70tps on TP4 vs 262k ctx at 40tps on DCP4.

Builders reported a flip — Codex/GPT-5.5 now wins on UI generation and copywriting, while Opus is relegated to strategy and creative writing. @tounano shared the standout workflow: generate UI mockups (including hero shots) with GPT-Image-2, then feed screenshots to Codex for ~90% accurate code. Iterating on images beats iterating on code and parallelizes across chat windows.

@bartekadamczyk asked how builders orchestrate multiple agents and manage workflow memory. anonymous runs one main agent in Telegram delegating to others via group chat, with Jira ticketing, GH Actions CI gates, and a custom control center for memory/context/billing. anonymous uses a Telegram ops agent with a GitHub Issues task repo. anonymous prefers Claude tmux dispatched to Telegram.

@iamgalba detailed his Google Ads scaling: 20-30 presell pages/month across listicles, advertorials, how-tos, and VSLs, each with 20 PMax image ads and A/B/anonymous PostHog tests; winners promoted to Demand Gen on Discover + YT in-feed. Separately, anonymous and @Uziarbiv discussed deepfake celebrity advertorials on Taboola/CBS placements causing CTR inflation legitimate buyers can'anonymous compete with.

Key Takeaways

  • Claude ID verification is rolling out — widely seen as government-driven and likely to cascade to other providers as they release more powerful models.
  • Self-hosted GLM 5.2 on 4-8 RTX 6000 Pros delivers near-Opus 4.8 coding at electricity cost; card prices climbing fast (750k→850k rub in 3 months).
  • New UI stack: generate mockups with GPT-Image-2 first, then have Codex convert screenshots to code — faster and more parallelizable than prompting code directly.
  • Working agent orchestration: one ops agent in Telegram delegating to specialists, GitHub Issues or Jira for task tracking, custom control center for memory/context/billing.
  • For AI trading or any prompted system, the evaluation harness for accepting/rejecting outputs matters more than the model or the prompt itself.

Hot Threads

anonymousstarted

Claude ID verification announcement and implications

20 replies9 participants
@tounanostarted

GPT-Image-2 + Codex as the new UI generation stack

18 replies4 participants
anonymousstarted

Deepfake celeb advertorials inflating CTRs across native

16 replies3 participants

Linked Items